PDA

View Full Version : Who is Dean Oliver?



DenButsu
10-12-2009, 12:32 PM
Among other things, he's the guy who says this:


In basketball, Oliver has "the four factors" he regards as the holy grail turnovers per possession, offensive rebounding percentage, free throws made per field goals attempted and effective field-goal percentage (which gives 50 percent more credit to 3-point shots than normal field-goal percentage).

"If you can control those four things offensively and defensively you win," he said.Denver Post (http://www.denverpost.com/nuggets/ci_13541239?source=rss)



More simply put, he's the Denver Nuggets stats man. Which for most NBA fans means he's some dude we don't know about. But among stats geeks his name is well known. He's written a bunch of books. And he's one of the basketball minds who has changed the way a lot of people think about the game. So if that's the kind of thing you care about, you should check that link, and then dig in deeper into some of his work.

I can't claim to have done so too much (I don't own any of his books, and haven't read them). But his rep is pretty well established, from what I understand.

JordansBulls
10-12-2009, 12:49 PM
The formulas are good because it helps to show who really has had the best stats. Like how do we know whose line is better if a big man gets:

22 ppg / 12 rpg / 3 apg / 3.5 bpg / 1.5 spg / 2.5 tpg / 52% fg

vs

A Guard who gets

28 ppg / 5 rpg / 5 apg / 0.8 bpg / 2.2 spg / 3.0 tpg / 47% FG

Chronz
10-12-2009, 01:10 PM
Thats not his thing JB, he doesnt like the idea of a player contributions being boiled down into one # (Though he readily admits PER/WARP is the most consistent/telling) and only describes his stats as performance evaluators.

Dean O was the guy who made per possession metrics "famous", hes one part of the holy trinity of stat rockstars. With a bunch of disciples and up and comers beneath him.

You guys are lucky to have him, but he got basis of his stuff from old school coaches who always talked in terms of possessions being like currency. He along with Rosenbaum and to a certain degree Hollinger completely changed the way I valued stats. I can safely say without them Id still be the moron that I was years ago. Looking back its ridiculous at some of the things I argued.

theuuord
10-12-2009, 01:13 PM
When did he move to the Nuggets? Last I heard he was on the Thunder (or, to be more precise, the Sonics).

Hustla23
10-12-2009, 01:24 PM
I'm interested in his work.
Chronz or theuuord, do you guys know what kind of mathematical background you need to comprehend his formulas and theories ?

The highest crap I've taken is calculus but I plan to take linear algebra either next term or next year.

theuuord
10-12-2009, 01:36 PM
I'm interested in his work.
Chronz or theuuord, do you guys know what kind of mathematical background you need to comprehend his formulas and theories ?

The highest crap I've taken is calculus but I plan to take linear algebra either next term or next year.

most of it is a solid statistics background, for instance if you know how to do stuff in excel you should be fine. that being said, the offensive/defensive ratings are way more complicated than that. that being said, he's actually surprisingly good at explaining most of his theories - I bought Basketball on Paper before I actually knew the math to understand ORtg and I still got the concept pretty well.
I highly suggest the book - it's really eye-opening, and it's also well-written - it never really got boring, even during the statistical explanations.

Chronz
10-12-2009, 01:41 PM
When did he move to the Nuggets? Last I heard he was on the Thunder (or, to be more precise, the Sonics).

Im pretty sure that was Pelton

theuuord
10-12-2009, 01:44 PM
Im pretty sure that was Pelton

Funny thing is, we're both right. I double-checked, Oliver was on the Sonics til 05, and then the Nuggets grabbed him. Now Pelton is with the Zombie Sonics.

Hustla23
10-12-2009, 01:46 PM
most of it is a solid statistics background, for instance if you know how to do stuff in excel you should be fine. that being said, the offensive/defensive ratings are way more complicated than that. that being said, he's actually surprisingly good at explaining most of his theories - I bought Basketball on Paper before I actually knew the math to understand ORtg and I still got the concept pretty well.
I highly suggest the book - it's really eye-opening, and it's also well-written - it never really got boring, even during the statistical explanations.
Thanks man.

Do you need to take statistics to really get it?

I really have no interest in statistics whatsoever lol.

Chronz
10-12-2009, 01:48 PM
I'm interested in his work.
Chronz or theuuord, do you guys know what kind of mathematical background you need to comprehend his formulas and theories ?

The highest crap I've taken is calculus but I plan to take linear algebra either next term or next year.
Some of his work is pretty complex especially the team ratings across different eras but his individual offensive stuff is pretty basic.

If you know how to evaluate running backs in football, you'll understand his offensive stats. Basically its like, yards per carry (efficiency/offensive rating) and the amount of carries (touches/usage) that define how good of an offensive player you are. There are variable that can effect this, just like in football how having a good offensive line will help your ratings, but its easier to sort out in basketball.

His defensive stuff is messy and sometimes not very accurate (he'll admit this) but it makes for an important starting point.

theuuord
10-12-2009, 01:49 PM
Thanks man.

Do you need to take statistics to really get it?

I really have no interest in statistics whatsoever lol.

Yes and no. I didn't take stats until college but I was always messing with Excel in HS in my free time, so I knew most of what he was talking about. Either way, I'd suggest getting the book - his writing is very accessible, even if you have trouble digging into the statistical side of it.

Hustla23
10-12-2009, 01:58 PM
Alright, thanks fellas.